/ . HoTt. Sci. 
Vol. 1 (1): 61-63, 2006 

Influence of morphological characters on the yield of apricot 
{Prunus armeniaca L.) - A statistical approach 

P. K. Mahajan and M. Thakur 
Department of Basic Sciences 

Dr. Y. S. Parmar University of Horticulture & Forestry 
Nauni (Solan) 173230, India 

E-mail:pawan_uhf@yahoo.com 

ABSTRACT 

Discriminate Analysis was carried out to formulate the categorization rule for allocating the apricot tree to 
"High Yielder Group" and "Low Yielder Group". Factor Analysis method was also applied to extract the basic 
factors underlying the observed morphological characters of apricot for both the High and Low Yielder groups. 
The study brought out five basic factors explaining 69.35% of the total variation in the case of High Yielder 
population and six factors explaining 74.14% of the total variation in the case of Low Yielder population, 
respectively. The first factor in both the populations contains the same variables viz. stem girth, number of 
branches and leaf area which indicate that these variables play an appreciable/significant role in increasing the 
yield of apricot (21 % in the case of High Yielder and 16.33% In the case of Low Yielder population). 

Key words: Apricot, morphological character, yield 

INTRODUCTION 

Apricot {Prunus armeniaca L.) is an important fruit 
crop of temperate regions of the world. In India, it is one of 
the most remunerative fruit crops cultivated in the mid hill 
zone of Himachal Pradesh. For improvement in productivity 
of a crop having enriched qualitative properties, the genetic 
selection of desirable traits is of utmost importance. Impact 
of morphological characters of a fruit crop on yield is 
desirable for its crop improvement programme and this can 
be achieved through multivariate statistical techniques. 
Moore (1965), for the first time, reported the use of 
multivariate analysis for quantifying yield component 
interactions in a horticultural crop. Ramachander et al 
(1979) used Factor Analysis in onion and reported two basic 
factors representing indices of plant vigour and flowering 
responsible for increasing the yield of onion. Schrevens et 
al (1995) used principal component analysis. Factor 
Analysis and biplots to characterize the quality evaluation 
of tomatoes during shelf life in relation to specific 
treatments. In this paper, an attempt has been made to bring 
out the basic factors (linear combination of morphological 
characters) contributing significantly towards the yield by 
using discriminant and factor analyses. The techniques are 
described in detail in standard statistical books such as 
Anderson (1958) and Kendal and Stuart (1968). 

MATERIAL AND METHODS 

The data for the present investigation were taken 
on the 'New Castle' variety of Apricot, growing in the 
research farm of the department of Pomology of the Dr Y S 
Parmar University, Solan during the year 2003. The 
optimum sample size (n = 30) of trees was determined by 
following a two step approach proposed by Stein (1945) 
and Cox (1952). Thereafter, thirty apricot trees were 
selected using simple random sampling technique (without 
replacement) from an orchard having one hundred trees of 
25 years old plantation. From each of these thirty trees, 
four branches were chosen randomly from each of the four 
directions as per the practice in vogue and observations on 
the following characters were recorded: 

Xj: Number of spurs per branch 

X^: Length of spurs (cm) 

X3 : Number of flowers 
per branch 

X^: Number of fruits per branch 

Xj : Fruit weight (g) 

X^: Stem girth (cm) 

Xg: Height of tree 

X^: Number of 
branches 

Xjg : Leaf area (cm) 

X : Spread of tree 

Xg: Annual shoot extension growth (cm) 

Discriminate analysis was carried out to define a 

mailto:pawan_uhf@yahoo.com


Mahajan and Thakur 

systematic and statistically valid procedure for categorizing 
the trees as 'low' and 'high' yielder. To bring out the basic 
factors associated with the above referred morphological 
characters of apricot, the data of two populations-High 
Yielder (Population I) and Low Yielder (Population II)-were 
subjected to factor analysis. 

RESULTS AND DISCUSSION 

Observations were first divided into two groups on 
the basis of previous year's data. Thereafter, a discriminant 
function was fitted by considering the above referred eleven 
characteristics and was found to be 

D = -5.023+0.648X,+0.016X, 
4 5 

This equation reveals that the characters of fruit 
weight (Xj) and number of fruits (X )̂ play a significant 
role to discriminate the two groups. To test the statistical 
hypothesis of no difference in mean vectors {[i^ and î̂ ) of 
eleven characters for these two groups, the value of Wilk's 
lambda (A) was obtained to be 0.214. In turn, the computed 
value of chi-square (x^) was 151.887 and hence the 
hypothesis of equality of group mean vectors was rejected. 
Having found that the groups differ statistically, the 

Table 1. Rotated Factor Matrix - Population I (High Yielder) 

individuals/trees were assigned to group I (High Yielder) if 
D > m otherwise to group II (Low Yielder), where m=0.359 
is the average of groups centroids. The groups formed on 
the basis of this allocation rule were subjected to factor 
analysis and population wise the results are discussed 
below: 

Population I (High Yielder) 

The rotated factor matrix and communalities are 
given in Table 1. This table reveals that the first five factors 
be retained and the sixth factor corresponding to an eigen 
value X = 0.945 is ignored (Guttman's lower bound principle 
according to which any ^<1 should be ignored). Ignoring 
the non-significant correlations, the orthogonal factors 
extracted can be expressed as: 

Factor 

F, = 0 . 7 5 X 5 + 0 . 6 8 X ; Q + 0 . 6 1 X , 

F, = 0 . 7 6 X , + 0 . 7 5 X , 

F3=0.71X,+0.57Xg_+0.51X5 

F^ =0.71X2+0.45X7 

F5 = 0.58X3+0.39X,+0.35X5 

Variance explained 
(% of total) 

21.00 

14.91 

13.90 

09.81 

09.73 

Variables 
Number of spurs (X,) 
Length of spur (X )̂ 
No. of flowers per branch (Xj) 
No. of fruits per branch (X )̂ 
Fruit weight (X^) 
Shoot extension (X )̂ 
Stem girth (X,) 
Height of tree (X^) 
No. of branches (X,) 
Leaf area (X^^) 
Spread of tree (X,,) 
Eigen values 

F. 
0.545 

-0.350 
-0.354 
0.234 

-0.088 
0.074 
0.613 

-0.350 
0.748 
0.682 
0.412 
2.310 

*Variable's highest loading is underlined 

Table 2. Rotated Factor Matrix 

Variables 
Number of spurs (X,) 
Length of spur (X )̂ 
No. of flowers per branch (X3) 
No. of fruits per branch (X^) 
Fruit weight (Xj) 
Shoot extension (X^) 
Stem girth (X,) 
Height of tree (X^) 
No. of branches (X,) 
Leaf area (X ĵ) 
Spread of tree (X^^) 
Eigen values 

F. 
0.383 
0.243 

-0.276 
0.066 
0.531 
0.763 

-0.023 
-0.465 
0.748 

-0.181 
-0.486 
1.640 

F, 
0.042 

-0.205 
-0.273 
0.719 
0.512 
0.197 
0.043 
0.571 

-0.346 
-0.343 
0.376 

1529.000 

- Population II (Low Yielder) 

F. 
0.316 

-0.194 
-0.158 
0.239 
0.105 

-0.162 
0.715 
0.326 
0.544 
0.781 

-0.118 
1.796 

F, 
0.439 

-0.018 
0.435 

-0.099 
-0.408 
0.100 
0.404 
0.583 

-0.507 
0.201 

-0.252 
1.436 

F, 
0.350 
0.248 

-0.113 
0.700 
0.502 

-0.033 
-0.106 
0.402 

-0.100 
0.044 
0.406 
1.396 

F. 
-0.377 
0.706 

-0.152 
0.267 

-0.086 
-0.054 
0.455 

-0.177 
-0.220 
0.214 

-0.007 
1.079 

F. 
-0.346 
0.533 

-0.328 
-0.060 
0.333 
0.504 
0.430 
0.120 

-0.168 
-0.025 
-0.410 
1.355 

F, 
-0.389 
-0.179 
0.575 
0.385 
0.348 
0.014 

-0.203 
-0.378 
0.175 
0.259 

-0.005 
1.069 

F, 
-0.371 
-0.445 
0.388 
0.362 
0.269 
0.142 

-0.139 
0.298 
0.246 
0.007 

-0.299 
1.122 

F. 
0.272 
0.348 
0.108 

-0.099 
0.388 

-0.395 
-0.091 
-0.229 
0.060 

-0.339 
0.490 
0.945 

F. 
0.118 

-0.450 
-0.018 
-0.197 
0.144 
0.767 

-0.031 
-0.078 
-0.074 
0.142 
0.388 
1.051 

Communalities 
0.833 
0.840 
0.930 
0.915 
0.879 
0.847 
0.713 
0.906 
0.875 
0.789 
0.805 

Communalities 
0.703 
0.892 
0.919 
0.842 
0.838 
0.911 
0.892 
0.794 
0.660 
0.759 
0.880 

•Variable's highest loading is underlined 

/ Hon. Sci. 
Vol. 1 (1): 61-63, 2006 

62 



Morphological characters on the yield of apricot 

In the present case, the first factor (F,) is a 
combination of number of branches (X^), leaf area (Xĵ ) 
and stem girth (X^). This factor signifies Plant Vigor, which 
indicates the general health of the plant. The second factor 
(F )̂ is the combination of the annual shoot extension growth 
(Xg) and number of branches (X^). If we ignore the relative 
low weighting 0.53 of fruit weight, the second factor (F )̂ 
signifies Plant Growth. The third factor (F3) is the 
combination of number of fruits (X^), height of tree (X-^) 
and fruit weight (X^). This factor signifies Yield Factor. 
The fourth factor (F )̂ is the combination of length of spurs 
(Xj) and stem girth (X^). This factor signifies Volume and 
Spur of plant and the fifth factor (F^) is combination of 
fruit weight (X^), number of fruits per branches (X )̂ and 
number of flowers per branches (X3). This factor may be 
regarded as Fruitfiilness or Fruiting. 

Population II (Low Yielder) 

As per eigen values (Table 2), six factors extracted 
along with the contributing variances are given below: 

Factors Variance explained 
(% of total) 

16.33 

13.05 

12.69 

12.32 

10.20 

09.55 

F, = 0.78X,„+0.72X^+0.54X5 

F^ = 0.58Xg+0.44X,+0.43X3 

F, = 0.70X,+0.50X, 
3 4 5 

F, = 0.53X,+0.50X,+0.43X, 
4 2 6 7 

R = 0.39X,+0.36X, 
5 3 4 

F = 0.77X,+0.39X„ 
o o 11 

In this case, first factor (Fj) is the combination of 
leaf area (X^ )̂, stem girth (X )̂ and number of branches 
(X(,). This factor signifies the Plant Vigour, which indicates 
the general health of a plant. The second factor (F )̂ may be 
interpreted as the Plant Vigour and Fruitfulness factor. Third 
factor (Fj) is the combination of the number of fruits (X )̂ 
and weight of fruits (X^). This factor signifies the Yield 
Factor. Fourth factor (F )̂ is the combination of length of 
spurs (Xj), annual shoot extension growth (X^), stem girth 
(X )̂ and spread of tree (Xj,). This factor signifies the Plant 
Growth. Fifth factor (F,) is the combination of the number 

of flower per branch (X3) and number of fruits (X^). It refers 
to the Fruitfulness or Fruiting. Sixth factor (F^) is the 
combination of annual extension growth (X )̂ and spread 
of tree (Xj,). This factor again signifies the Plant Growth. 

Thus, the Factor Analysis has brought out the basic 
factors (after ignoring non significant correlations at 5% 
level of significance) associated with the morphological 
characters of the apricot for low yielder and high yielder 
populations. It is evident from the results that in the case of 
high yielder population stem girth, number of branches and 
leaf area are the variables that form part of the first factor. 
This factor accounts for 21 % of the total variation towards 
yield. The same variables were found to be responsible in 
forming the first factor which contributes 16.33% of the 
total variation towards yield in the case of low yield 
population. Similar inference can be drawn from the 
remaining factors for both populations as detailed in the 
population - wise discussion. It may be added that without 
resolution of the rotated orthogonal matrix to oblique axis, 
it would not have been possible to bring out the meaningful 
factors for apricot. 

REFERENCES 

Anderson, T.W. 1958. An Introduction to multivariate 
statistical analysis. John Wiley & Co., London. 374p. 

Cox, D.R. 1952. Estimation of double sampling, Biometrika, 
39:217-27. 

Kendal, M.G. and Stuart, A. 1968. The Advanced theory of 
statistics. Vol. III. Charles Griffin. 

Moore, C.S. 1965. Inter-relations of growth and cropping 
in apple trees studied by the method of component 
analysis. J. Hort. Sci.,. 40:133-49. 

Ramchander, RR., Biswas, S.R., Singh, D.P. and Pathak, 
C.S. 1979. Factor analysis in onion (Allium cepa L.). 
Curr. Sd.. 48:137. 

Schrevens, E., Neyens, K., Verreydt, J., Baerdemaeker, 
J.D.E., and Portier, K. 1995. Quality evaluation of 
fruit by means of multivariate analysis on non-
destructive parameters. Acta Hort., 379:569-578. 

Stein, C. 1945. A two sample test for a linear hypothesis 
whose power is independent of the variance. Ann. 
Math. Statist., 16:243-258. 

(MS Received 15 March 2006 Revised 14 July 2006) 

J. Hon Sci. 
Vol. 1(1); 61-63, 2006 

63